Measuring the Willingness to Pay for Fresh Water Cave Diving

24
1 Measuring the Willingness to Pay for Fresh Water Cave Diving William L. Huth 1 , O. Ashton Morgan 2 1 Professor University of West Florida Department of Marketing and Economics 11000 University Parkway Pensacola, FL 32514 USA Tel: (850) 474 2826 Email: [email protected] 2 Assistant Professor Department of Economics 3094 Raley Hall Appalachian State University Boone, NC 28608 USA Email: [email protected] Tel: (828) 262 2927 Fax: 00 1 (828) 262 6105 JEL Codes: Q26; Q51 JEL Keywords: Contingent Valuation Model; Willingness to Pay; Cave Diving; Scope Sensitivity

Transcript of Measuring the Willingness to Pay for Fresh Water Cave Diving

Page 1: Measuring the Willingness to Pay for Fresh Water Cave Diving

1  

Measuring the Willingness to Pay for Fresh Water Cave Diving 

 

William L. Huth1, O. Ashton Morgan2 

 

 1 Professor 

University of West Florida 

Department of Marketing and Economics 

11000 University Parkway 

Pensacola, FL 32514 

USA 

Tel: (850) 474 2826 

Email: [email protected] 

 

 2 Assistant Professor 

Department of Economics 

3094 Raley Hall 

Appalachian State University 

Boone, NC 28608 

USA 

Email: [email protected] 

Tel: (828) 262 2927 

Fax: 00 1 (828) 262 6105 

 

 

JEL Codes: Q26; Q51 

 

JEL Keywords:  Contingent Valuation Model; Willingness to Pay; Cave Diving; Scope Sensitivity 

 

 

 

 

 

 

 

Page 2: Measuring the Willingness to Pay for Fresh Water Cave Diving

2  

Abstract 

Fresh water springs are unique natural resources that are contained within public lands across 

the United States. Natural resource management on public lands generates many interesting 

policy issues as the competing goals of conservation, recreational opportunity provision, and 

revenue generation often clash. As demand for recreational cave diving sites increases, the 

paper provides natural resource site managers with the first statistical estimate of divers’ 

willingness to pay to dive fresh water cave and cavern systems. Using a contingent valuation 

model approach and correcting for hypothetical bias, we find that divers’ median willingness to 

pay for cave diving opportunities at the site of interest is approximately $68 per dive. Model 

results also provide evidence of diver sensitivity with respect to scope as individuals are willing 

to pay more for dives that are higher in quality.  

 

1. Introduction 

A unique natural resource in the state of Florida is the number and size of fresh water 

springs. The Florida Geological Survey has inventoried more than 700 springs, of which 

33 are considered first magnitude, having an average flow of 100 cubic feet per second 

(2.83 cubic meters per second) or more. The concentration of springs in Florida is not 

duplicated anywhere else on the earth. The Florida Department of Environmental 

Protection (DEP) has the management responsibility for Florida’s public lands and the 

Division of Recreation and Parks manages a system of 160 State Parks that combine to 

put 700,000 acres scattered throughout the state under public management (Florida, 

Page 3: Measuring the Willingness to Pay for Fresh Water Cave Diving

3  

DEP, 2009). Approximately 70% of Florida’s Parks are related in some way to a natural 

spring. Natural resource management on public lands generates many interesting 

policy issues. The competing goals of conservation, recreational opportunity provision, 

and revenue generation often clash. In recent times, as public sector budgets have 

shrunk, natural resource managers have also been forced to search for revenue 

generation alternatives to supplement shrinking budgets. This situation is especially 

true in Florida.  

For federal funding purposes states are required to publish recreation plans every 5 

years. The most recent plan for Florida was produced by the Florida DEP in 2007. 

Chapter 5 of the plan addresses “outdoor demand and need” and it is stated that “Since 

outdoor recreation resources and facilities are generally felt to be ‘free’ goods and 

services, ‘demand,’ as an economic concept, does not lend itself to practical 

application,“ (DEP, 2009).  What is provided below is a step towards providing a 

mechanism for practical application of demand measurement for a resource with public 

good elements. Because springs are an important natural resource in Florida and a key 

element in many state parks, a spring based state park was selected for the contingent 

valuation demand modeling that follows. 

 

 

Page 4: Measuring the Willingness to Pay for Fresh Water Cave Diving

4  

2. Site of Interest 

The Edward Ball Wakulla Springs State Park is located just south of Tallahassee Florida 

in the Woodville Karst Plain (WKP). This area is well known for its karst, or landforms 

that have been modified by dissolution of soluble rock (eg. limestone), resulting in a 

terrain that is characterized by natural springs, sinkholes, sinking and rising streams, 

and caves. Wakulla Spring is the park’s centerpiece feature and this particular spring is 

considered world class with regard to its flow and the cave system that channels its 

flow. In 2007, after years of exploratory effort, divers finally connected a number of 

other systems in the WKP to Wakulla. They entered at Turner Sink in the Leon Sinks 

Cave System and surfaced over 20 hours (a 6.5 hour dive with 14 hours of 

decompression due to 300 feet dive depth) later at Wakulla Spring after following 

almost 7 miles of cave passage. This established the Wakulla‐Leon Sinks Cave System as 

the longest underwater cave in the U.S. (Kernagis et al. 2008). While Wakulla Spring is 

the most prominent feature in the park, it does contain other springs as well, including 

Sally Ward and Emerald Spring that also have associated cave systems.  

Bonn and Bell (2003) measured the economic impact from Wakulla Spring along with 

the impact from three other springs in Florida (Ichetucknee, Volusia, and Homasassa 

Springs). Although this set of four springs is not a representative sample of all Florida 

springs, Bonn and Bell (2003) concluded from their visitor surveys that for a “typical 

Page 5: Measuring the Willingness to Pay for Fresh Water Cave Diving

5  

spring” annual aggregate visitor spending is marginally in excess of $17 million. They 

also noted that visitors to the springs average about $46 per day in spending, and while 

they did not distinguish between visitor recreational purposes, they did note that visitor 

spending varies significantly by spring. For Wakulla Spring they found an annual direct 

economic impact of $22.2 million on Wakulla County wherein the spring is located. 

They also indicated that some 180,793 visitors came to the spring in 2002 spending 

about $90 per day, 70% of whom were from outside the county.1  

3. Background and the Contingent Valuation Model 

We develop the first contingent valuation model (CVM) study measuring the economic 

benefit associated with cave diving. In the contingent valuation/travel cost literature, 

some attention has been directed at valuing recreational diving. One group of diving 

studies considered the economic benefits associated with diving Marine Protected 

Areas (MPAs).  Quantifying the economic benefits from diving MPAs can provide local 

conservation and management groups with important policy‐based feedback to 

determine whether access fees are an appropriate method to fund MPAs and help 

conserve and protect the natural ecosystems (Arin and Kramer, 2002; Hall et al., 2002; 

van Beukering et al., 2004; Barker and Roberts, 2004). Another group of studies focused 

                                                            1 The visitation estimate included individuals from Leon County. Leon County, which includes the 

Tallahassee metro area, borders Wakulla County. As Wakulla Spring is a 20 minute drive south of 

Tallahassee, including Leon County visitors as external might well be overestimating the annual 

visitation economic impact associated with the spring.   

Page 6: Measuring the Willingness to Pay for Fresh Water Cave Diving

6  

on diving natural and artificial reefs. Using both CVM and travel cost models (TCM), 

these studies found significant use values associated with diving reefs. For example, 

Ditton et al. (2001) and McGinnis et al. (2001) calculated divers’ annual willingness to 

pay (WTP) for recreational reef diving off the coasts of Texas and California 

respectively, with estimates ranging from $205 to $646 per year depending on the 

disclosure mechanism used in the CVM framework. Also, Kragt et al. (2009) used panel 

data to measure the value of dive trips to the Great Barrier Reef off the Australian coast 

and calculated consumer surplus estimates of approximately $150 per trip. To further 

value the potential spillover benefits of diving artificial reefs, Leeworthy et al. (2006) 

considered whether deployment of artificial reefs could alleviate the diving pressure on 

adjacent natural reefs. They found a 13.7% decrease in use of surrounding natural reefs 

following the sinking of the USS Spiegel Grove as an artificial reef off Key Largo in the 

Florida Keys. Most recently, Morgan et al. (2009) used a TCM approach to value 

recreational diving on the USS Oriskany (the World’s largest artificial reef). Results 

from different model specifications indicated per‐person, per‐trip use values between 

$480 and $750. In addition, they measured the value of “bundling” a second vessel 

alongside the Oriskany to create a multiple‐ship reefing area between $220 and $1,160 

per diver, per year. However, despite the contribution of research directed at valuing 

recreational diving, and the growing participation in the sport, no‐one has considered 

the economic value associated with cave diving. In the U.S. alone, there are hundreds of 

Page 7: Measuring the Willingness to Pay for Fresh Water Cave Diving

7  

cave diving sites and a cave diver population, based on association memberships and 

training records, consisting of thousands of divers. Based on the increase in demand for 

cave/cavern diving sites and the need for resource managers to offset budget constraints 

with new streams of revenue generation, our results will provide public managers with 

valuable statistical feedback on the use values and potential economic efficiency 

associated with fresh water cave diving within the state park system.  

We also consider scope effects by measuring divers’ WTP for diving cave and cavern 

systems that vary in diver experience requirement and dive quality. Essentially, WTP 

should be non‐decreasing in scope. In a CVM framework, scope sensitivity exists if 

respondents’ WTP for a public good of greater quantity or quality is significantly 

different. A priori, divers would be expected to exhibit a higher WTP for a more 

advanced cave dive that goes beyond the ambient light zone and penetrates the cave 

relative to a cavern dive that does not go beyond the cave entrance area. In the 

economic literature, findings on scope effects remain mixed. Some previous research 

found scope insensitivity effects, meaning that respondents are not willing to pay more 

for an increase in quantity or quality of the public good (Schkade and Payne, 1994; 

Whitehead and Finney, 2003; Whitehead, 2005). Others have found that WTP estimates 

are sensitive to the scope of the policy (Carson, 1997; Powe and Bateman, 2004; Morgan 

et al., 2009). Finally, some research has argued that a test for scope effects is a test of the 

validity of the CVM framework with scope insensitivity suggesting that the CVM 

Page 8: Measuring the Willingness to Pay for Fresh Water Cave Diving

8  

method would not be valid for policy analysis (Diamond and Hausman, 1994).   

 

4. The Survey 

A CVM survey was developed to elicit divers’ socio‐demographic details and their WTP for two 

different cave dives and a cavern dive that are currently closed to anything but permit‐based 

dives. A portion of the survey was pre‐tested on 46 respondents at the 2008 Cave Diving Section 

of the National Speleological Society annual meeting in Marianna, FL.  For the study the 

population of interest was individuals known to have dived cave systems similar to that which 

would be available at Wakulla State Park should it be opened for recreational diving. Diver 

registrations at a nearby cave system (Jackson Blue in Marianna Florida) were used so that 

divers with the requisite skill sets who had actually been in the area (by vehicle Jackson Blue 

and Wakulla Springs are about one hour apart) were surveyed. Surveys were sent to 525 

individuals known to have dived in similar cave systems with a stamped addressed return 

envelope included to increase responses. Also to increase the response rate, we informed 

potential respondents that they would be entered into a random draw for one of three $100 

vouchers at a local dive shop if they completed and returned a survey. 

Insert Table 1 here 

 

There were 146 responses received giving a response rate of 27.8%. The average age of 

respondents was 45.8 years, earning an annual income of $102,430, with a bachelor’s degree. 

The majority of respondents were male (87%) and married (74%). The average diver in the 

Page 9: Measuring the Willingness to Pay for Fresh Water Cave Diving

9  

sample had a full cave certification level and would incur $679 in travel costs to access the 

Wakulla site.2  

 

Insert Table 2 here 

5. Estimation Methodology 

Consider a diver who receives utility, u, from a dive site quality measure, q, and a composite of 

all other goods, z. Each diver chooses to maximize utility, u(q, z), subject to a budget constraint, 

y=pz, where y is diver income and the price of z is normalized to one. Solving for the indirect 

utility function gives v(q, y). A diver’s willingness to pay (WTP) to dive a site is given by the 

monetary payment that equates indirect utility, such that v(q0, y) = v(q1, y – WTP), where q0 is the 

current site quality, and q1 is the new site quality. As such, WTP represents the amount of 

money that must be subtracted from respondent’s income at the higher level of site quality to 

leave diver utility unchanged.   

An issue that arises with asking respondents questions regarding their anticipated behavior 

based on a hypothetical scenario is the level of certainty in their responses. Critics of CVM and 

WTP values dispute whether respondents’ stated WTP estimates approximate their true WTP. 

Diamond and Hausman (1994) argued that stated preference responses to hypothetic scenarios 

do not correspond to what the individual would pay in real life, and suggested that responses 

would be less if the respondent had to actually pay for the provision at that point in time. 

                                                            2 Travel costs are calculated as round trip travel expenses, plus site fees, plus the opportunity cost of time 

estimates. Round trip distance was estimated using the PC*Miler software. Per‐mile travel costs were 

assumed to be $0.48. The opportunity cost of time for the roundtrip travel was calculated as one‐third the 

hourly wage foregone assuming the average diver sampled works 2,080 hours per year. 

Page 10: Measuring the Willingness to Pay for Fresh Water Cave Diving

10  

Kahneman and Knetsch (1992) further contended that, as individuals yield satisfaction from 

stating that they will contribute to a cause without actually having to pay, CVM valuations 

merely reflect individuals’ WTP for moral satisfaction, and as such, are not good estimators of 

their true WTP.  

To counter the criticisms of CVM methods, recommendations regarding survey design have 

been suggested to improve the validity of individual responses. One recent concept that has 

been introduced into the CVM literature is cheap talk. The notion of cheap talk was introduced 

as a means to mitigate hypothetical bias in an individual’s response. Including responses from 

individuals that are uncertain about the likelihood of actually paying the fee in a real situation 

can result in an overestimation of true WTP (Whitehead and Cherry, 2007). However, findings 

from experimental research have indicated that controlling for cheap talk in a CVM model 

removes hypothetical bias and provides WTP estimates that more closely approximate an 

individuals’ actual WTP (Cummings and Taylor, 1999).  

 

In the CVM framework developed here, three separate questions were asked to elicit divers’ 

WTP for new cave and cavern dives with varying quality levels. Each scenario represented a 

dive that is currently closed to anything but permit‐based dives but could be opened to the 

diving public with a Park policy decision. Similar caves in size and depth are currently open on 

both state and private land throughout Florida. The first scenario involved is a dive at Sally 

Ward Spring. Specifically, respondents were told “Sally Ward Spring is located on the entrance 

road to Wakulla State Park just before the entrance station. Sally Ward is also known as 

“Numero Uno” because it is said to be the number one rated cave dive in Florida. Your guided 

Page 11: Measuring the Willingness to Pay for Fresh Water Cave Diving

11  

dive would be a staged swim to the Balcony entrance into the Cube Room (a gymnasium sized 

room), and then a circuit around that room and exit.”  

Scenario 2 involved a dive at Wakulla Spring. Under this scenario, respondents were informed 

that “This dive is a cavern dive that does not go past the ambient light zone and remains in 

front of the cave entrance at 160 feet.” 

Scenario 3 also involves a dive at Wakulla Spring. Here though, divers were told that “This dive 

is a time and/or penetration limited Tunnel A cave dive that goes into Tunnel A and then to the 

“grand canyon,” (approximately a 400 foot penetration and a max depth of 225 feet) and then 

on to and no further than the junction of Tunnel B (a 1,100 foot penetration and a max depth of 

270 feet).” 

Before each WTP question, respondents were told that studies have shown that when people 

are asked about whether they are willing to pay for goods, such as this one, they often say yes at 

the time they are surveyed, but later think that they should have said no. This can be for a good 

reason, as people later realize that this would take money away from other things that are 

important to them. Respondents were informed that when considering their willingness to pay 

for a permit to dive the system, they should think carefully about whether they really would 

prefer to pay for this permit, or would prefer to continue purchasing other things that are 

important to them.  

After each scenario, the respondent was then asked: 

Consider for a moment that to gain access for this dive, you will be asked to pay for a 

dive permit. Suppose that the price of the permit is $A, would you purchase it and thus 

be able to dive the cave/cavern? 

Page 12: Measuring the Willingness to Pay for Fresh Water Cave Diving

12  

In each case, $A is a randomly assigned permit price variable.3 Respondents were presented 

with three possible answers: yes, no, don’t know, where any don’t know response were 

categorized as a no response. These responses were used to estimate the full version of the 

model (Model 1). 

To account for potential hypothetical bias in individuals’ responses, after each WTP question, 

the individual was asked a follow‐on certainty statement. Here, each respondent was asked “on 

a scale of 1 to 10 where 1 is very uncertain and 10 is very certain, how certain are you that you 

would pay a $A license fee.” To control for hypothetical bias, a second set of models was run in 

which only responses from individuals that stated a certainty of 7 or more were included 

(Model 2).   

For both models, a logistic model specification is estimated with the probability of saying yes 

P(Yes) as the dependent variable. This can be written as: 

 

1/ 1 exp

                    (1) 

 

6. Results 

Before discussing the main results, analyzing the yes responses indicates that the divers 

sampled behaved in line with economic theory as an increase in annual license fees reduces the 

likelihood of a yes response.  

                                                            3 Permit fees were randomly assigned as $25, $50, $100, $200, or $300. 

Page 13: Measuring the Willingness to Pay for Fresh Water Cave Diving

13  

Insert Table 3 here 

For each scenario, there is a clear trend in the decline in the percentage of yes responses as the 

permit fee rises.  

Now considering the models, three different model specifications are run; one for each new 

cave/cavern dive scenario (Model 1). A second model that controls for hypothetical bias is also 

run under each specification (Model 2). Across all models, the log of the permit fee amount is 

negative and statistically significant at the 1% level, confirming that increased access fees reduce 

the likelihood of a respondent’s WTP. For all scenarios, having adjusted for hypothetical bias, 

the income variable is positive and significantly different from zero, indicating that cave diving 

is a normal good.  

Insert Table 4 here 

For  the most  part,  results  from  the  Sally Ward  and Wakulla Cave  dive models  are  similar. 

Intuitively, this makes sense as both dives represent more advanced cave dives that penetrate 

deeper into the cave systems, while the Wakulla cavern dive is a structurally different dive that 

does not venture beyond the ambient light zone or enter the cave itself. For the Sally Ward and 

Wakulla Cave dives, males are more likely to be willing to pay for a dive permit than females. 

Travel  costs are  important with  results  indicating  that  those  living  further  from  the  site with 

greater  travel  costs  are  less  likely  to  answer yes  to  the WTP question. Also,  including  travel 

costs to the closest substitute site indicates that those living further from the substitute site are 

more  likely  to  answer yes. For  the Sally Ward Cave dive, younger divers  are more  likely  to 

answer yes while married divers are more likely to say yes for the Wakulla Cave dive. Finally, 

certification level does not appear to be important in any model.  

Page 14: Measuring the Willingness to Pay for Fresh Water Cave Diving

14  

Respondents’ WTP and confidence intervals for each scenario are also estimated. There are a 

number of established methods to estimate WTP and confidence intervals in the CVM literature 

(such as the delta method and bootstrapping). Confidence intervals are estimated by simulation 

using the Krinsky‐Robb procedure (Habb and McConnell, 2002).   As the log of the permit fee 

amount (lnA) is also included in the model, the median WTP is estimated, with the mean WTP 

undefined (Haab and McConnell, 2002).  As such, estimates of respondents’ WTP are probably 

conservative as the median is lower than the mean (Groothuis et al., 2006).  

Insert Table 5 Here 

The median WTP estimates indicate that divers’ value the more advanced cave dives more than 

the cavern dive. Specifically, for the Sally Ward Cave dive, respondents’ median WTP is 

approximately $80, with a 95% confidence interval of from $56 up to $110. For the Wakulla 

Cave dive, respondents’ median WTP is approximately $84, with a 95% confidence interval of 

$50 to $128. For the cavern dive, WTP estimates fall to $34 with a 95% confidence interval of $15 

to $51. These results indicate that the sampled divers are sensitive to scope as they are willing to 

pay more for higher quality cave dives relative to the cavern dive. Based on a common criticism 

of CVM, this result also validates the policy‐based analysis approach of this research (Diamond 

and Hausman, 1994).  

When the estimates are corrected for hypothetical bias (Model 2), the WTP estimates decline. 

This was expected, as removing responses from divers that are not as certain of their answer 

provides a more conservative (and as has been argued, a more accurate) measure of 

individuals’ WTP. For the two proposed cave dives, WTP estimates fall to between $68 and $69 

Page 15: Measuring the Willingness to Pay for Fresh Water Cave Diving

15  

(with 95% confidence intervals between $35 and $106), while for the cavern dive, WTP declines 

to approximately $23 (with a 95% confidence interval between $4 and $41).   

 

7. Conclusion 

We developed a contingent valuation model to provide the first estimate of individuals’ WTP 

for fresh water cave diving. The results suggested that divers’ median WTP for these cave 

diving opportunities at Wakulla Springs is in the region of $68 per dive when controlling for 

hypothetical bias in responses. For cavern dives requiring less experience, WTP estimates are 

$23 per dive. As natural resource managers face shrinking budgets, our results indicated that 

recreational cave diving within state parks could provide an important revenue stream. At the 

aggregate level, based on the number of individuals that dive comparable sites in the region, we 

estimate that if the cave/cavern system was open to the public, it would attract approximately 

1,000 divers per year.4 Our sample diver population makes, on average, nine dives per year at 

the Jackson Blue site, so we use this number as an estimate of the annual number of expected 

trips. Using the assumed visitation rates, aggregate willingness to pay is approximately 

$612,000. Based on a 30‐year annuity with an assumed 5% yield, annual revenue generated 

would be $37,916. If the annual cost of managing the site with a new cave system open to the 

public is less than $37,916, then our results indicate that the site improvement represents an 

increase in economic efficiency.  

                                                            4 This may be considered an upper bound estimate as visitation depends on the diving restrictions that the park may 

impose on the public. We also assume that the majority of divers are interested in cave dives. 

Page 16: Measuring the Willingness to Pay for Fresh Water Cave Diving

16  

The results also provided evidence of diver sensitivity with respect to scope. That is, individuals 

are willing to pay more for dives that are higher in quality. This finding of scope sensitivity, 

together with model estimation results that conform to economic theory, indicated that the WTP 

estimates provide useful information for policy‐based analysis.  

 

   

Page 17: Measuring the Willingness to Pay for Fresh Water Cave Diving

17  

8. References 

Arin, T., Kramer, R.A., 2002. Divers’ Willingness to Pay to Visit Marine Sanctuaries: an 

Exploratory Study. Ocean and Coastal Management. 45(2),171–83. 

 

Barker, N.H.L., Roberts, C.M., 2004. Scuba Diver Behaviour and the Management of Diving 

Impacts on Coral Reefs. Biological Conservation. 120(4), 481–9. 

 

Bonn, M.A., Bell, F.W., 2003. Economic Impact of Selected Florida Springs on Surrounding Local 

Areas. Report for Florida Department of Environmental Protection. Division of State Lands, 

Florida Springs Task Force.  

Carson, R.T., 1997. Contingent Valuation Surveys and Tests of Insensitivity to Scope. In Kopp, 

R.J., Pommeerehne, W.W., Schwarz, N.  (Eds.), Determining  the Value of Non‐Market Goods: 

Economic,  Psychological,  and  Policy  Relevant  Aspects  of  Contingent  Valuation  Methods. 

London: Kluwer Academic Publishers. 

 

Cummings, R.G., Elliot, S., Harrison, G.W., Murphy, J., 1997. Are Hypothetical Refernda 

Incentive Compatible. Journal of Political Economy. 105, 609‐621. 

 

Diamond, P.A, Hausman, J.A., 1994. Contingent Valuation: Is Some Number Better than No 

Number? Journal of Economic Perspectives. 8(4), 45‐64. 

 

Ditton, R., Thailing, C., Reichers, R., Osburn, H., 2001. The Economic Impacts of Sport Divers 

Using Artificial Reefs in Texas Offshore Waters. Proceedings of the Annual Gulf and Caribbean 

Fisheries Institute. 54, 349‐360. 

 

Florida Department of Environmental Protection. 2009. Outdoor Recreation in Florida‐2007: 

Florida’s Statewide Comprehensive Outdoor Recreation Plan.  

Florida Department of Environmental Protection, Division of Recreation and Parks. 2007. 

Edward Ball Wakulla Springs State Park” Unit Management Plan.  

http://www.dep.state.fl.us/lands/files/lmr_reports/wakulla_2007.pdf. 

Groothuis, P.A, Groothuis, J.A., Whitehead, J.C., 2006. The Willingness to Pay to Remove 

Billboards and Improve Scenic Amenities. Journal of Environmental Management. 85(4), 1094‐

1100. 

 

Haab, T.C., McConnell, K.E., 2003. Valuing Environmental and Natural Resources: The 

Econometrics of Non‐market Valuation. Northampton, MA. Edward Elgar. 

  

Hall D.C., Hall, J.V., Murray, S.N., 2002. Contingent valuation of marine protected areas: 

Southern California rocky intertidal ecosystems. Natural Resource Modelling 15(3), 335–68. 

Page 18: Measuring the Willingness to Pay for Fresh Water Cave Diving

18  

 

Kahneman, D., Knetsch, J.L., 1992. Valuing Public Goods: The Purchase of Moral Satisfaction. 

Journal of Environmental Economics and Management. 22, 57‐70. 

Kernagis, D.N., McKinlay, C., Kincaid, T.R., 2008. Dive Logistics of the Turner to Wakulla Cave 

Traverse, in: Brueggerman, P., Pollock, N.W.,  (Eds.), Diving for Science 2008. Proceedings of the 

American Academy of Underwater Sciences Symposium. Dauphin Island, AL.  

Kragt, M.E., Roebeling, P.C., Ruijs, A., 2009. Effects of Great Barrier Reef Degradation on 

Recreational Reef‐trip Demand: A Contingent Behavior Approach. The Australian Journal of 

Agricultural and Resource Economics. 53(2), 213‐229. 

Leeworthy, V. R., Maher, T., Stone, E., 2006. Can Artificial Reefs Alter User Pressure on 

Adjacent Natural Reefs? Bulletin of Marine Science. 78(1), 29‐37. 

 

McGinnis, M., Fernadez, L., Pomeroy, C., 2001. The Politics, Economics, and Ecology of 

Decommissioning Offshore Oil and Gas Structures. Department of the Interior Mineral 

Management Services. 

 

Morgan, O.A., Massey, D.M., Huth, W.L., 2009. Diving Demand for Large Ship Artificial Reefs. 

Marine Resource Economics. 24, 43‐59. 

Nordlie, F.G., 1990. Rivers and Springs. In Myers, R.L., Ewel, J.J., (Eds.) Ecosystems of Florida, 

1990. Orlando: University of Central Florida Press. Pp. 392‐425. 

Powe, N.A., Bateman I.J., 2004. Investigating Insensitivity to Scope: A Split‐Sample Test of 

Perceived Scheme Realism. Land Economics 82(2), 258‐271. 

Schkade, D.A., Payne, J.W., 1994. How People Respond to Contingent Valuation Questions: A 

Verbal Protocol Analysis of Willingness to Pay for an Environmental Regulation. Journal of 

Environmental Economics and Management 26(1), 88‐109. 

van Beukering, P., Cesar, H., Dierking, J., Atkinson, S., 2004. Recreational Survey in Selected 

Marine Managed Areas in the Main Hawaiian Islands, in: Assessment of Economic Benefits and 

Costs of Marine Managed Areas in Hawaii. University of Hawaii; 2004. p. 14. Report for the 

Hawaii Coral Reef Initiative Research Program. 

 

Whitehead, J.C., Finney, S.S., 2003. Willingness to Pay for Submerged Maritime Cultural 

Resources. Journal of Cultural Economics. 27, 231‐240. 

 

Whitehead, J.C., 2005. Environmental Risk and Averting Behavior: Predictive Validity of Jointly 

Estimated Revealed and Stated Behavior Data. Environmental and Resource Economics. 32, 301‐

316. 

Page 19: Measuring the Willingness to Pay for Fresh Water Cave Diving

19  

Whitehead, J.C., Cherry, T.L., 2007. Mitigating the Hypothetical Bias of Willingness to Pay: A 

Comparison of Ex‐Ante and Ex‐Post Approaches. Resources and Energy Economics 29(4), 247‐

261. 

   

Page 20: Measuring the Willingness to Pay for Fresh Water Cave Diving

20  

Table 1. 

Variable Definitions 

 

Variable  Definition 

Age  Age of respondent in years 

Male  Dummy variable denoting respondent gender (male = 1, 0 

otherwise) 

Married  Dummy variable denoting respondent marital status (married = 1, 

0 otherwise) 

Income   Income of respondent ($1,000s) 

Educ  Respondent level of education 

Cert_lev  Cave certification level 

TC Site  Per person travel cost necessary for each respondent to dive at 

Jackson Blue 

TC Sub  Per person travel cost necessary for each respondent to dive at a 

substitute site (Ginnie Springs, FL) 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Page 21: Measuring the Willingness to Pay for Fresh Water Cave Diving

21  

Table 2.  

Summary Statistics (Obs = 146) 

 

Variable  Mean  Standard Dev.  Min  Max 

Age  45.79  10.36  18.00  66.00 

Male  0.87  0.34  0.00  1.00 

Married  0.74  0.44  0.00  1.00 

Income  $102.43  42.79  $25.00  $155.00 

Educ  4.64  3.33  0.00  6.00 

Cert_lev  3.72  0.80  0.00  4.00 

TC Site  $679.03  611.33  $45.30  $3,406.20 

TC Sub  $704.31  626.57  $52.50  $3,587.20 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Page 22: Measuring the Willingness to Pay for Fresh Water Cave Diving

22  

Table 3. 

Yes Responses by Payment Scenario 

 

  Response  $25  $50  $100  $200  $300 

Sally 

Ward 

Yes  17  19  13  11  6 

Total  25  25  31  32  31 

Percent  68%  76%  42%  34%  19% 

Wakulla 

Cavern 

Dive 

Yes  14  7  4  7  5 

Total  25  25  31  32  31 

Percent  56%  28%  13%  22%  16% 

Wakulla 

Cave Dive 

Yes  15  18  16  13  6 

Total  25  25  31  32  31 

Percent  60%  72%  52%  41%  19% 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Page 23: Measuring the Willingness to Pay for Fresh Water Cave Diving

23  

Table 4.  

Determinants of WTP for New Divesa 

 

  Sally Ward  Wakulla Cavern Dive  Wakulla Cave Dive 

Variable  Model 1  Model 2  Model 1  Model 2  Model 1  Model 2 

Intercept  4.56*** 

(1.71) 

3.77*** 

(1.81) 

3.10*** 

(1.83) 

0.85 

(2.02) 

1.83 

(1.62) 

1.48 

(1.71) 

 

Log Price  ‐1.34*** 

(0.28) 

‐1.39*** 

(0.29) 

‐1.22*** 

(0.29) 

‐1.04*** 

(0.31) 

‐0.94*** 

(0.24) 

‐0.92*** 

(0.25) 

 

Age  ‐0.03*** 

(0.02) 

‐0.03 

(0.02) 

‐0.00 

(0.02) 

0.01 

(0.02) 

‐0.02 

(0.02) 

‐0.02 

(0.02) 

 

Male  1.51** 

(0.72) 

1.63*** 

(0.79) 

1.27 

(0.83) 

1.38 

(0.92) 

1.22* 

(0.66) 

1.26* 

(0.72) 

 

Married  0.59 

(0.50) 

0.52 

(0.52) 

‐0.08 

(0.60) 

‐0.10 

(0.65) 

0.76* 

(0.46) 

0.81* 

(0.49) 

 

Income  0.00 

(0.00) 

0.00*** 

(0.00) 

0.02*** 

(0.01) 

0.02** 

(0.01) 

0.00 

(0.00) 

0.00*** 

(0.00) 

 

Cert. Level  0.16 

(0.24) 

0.31 

(0.28) 

‐0.21 

(0.25) 

‐0.05 

(0.30) 

0.27 

(0.24) 

0.30 

(0.26) 

 

Travel 

Cost Site 

‐0.01*** 

(0.00) 

‐0.01 

(0.00) 

‐0.01** 

(0.00) 

‐0.00 

(0.00) 

‐0.01** 

(0.00) 

‐0.01** 

(0.00) 

 

Travel 

Cost Sub 

0.01*** 

(0.00) 

0.01 

(0.00) 

0.00 

(0.00) 

0.00 

(0.00) 

0.01** 

0.00 

0.01** 

(0.00) 

 

Model χ2  47.68***  45.42***  34.73***  22.20***  33.19***  31.50*** a The top number in each cell is the estimated coefficient and the bottom number in 

parenthesis is the estimated standard error estimate. 

*** Significant at the p = 0.10 level 

** Significant at the p = 0.05 level 

* Significant at the p = 0.01 level 

 

Page 24: Measuring the Willingness to Pay for Fresh Water Cave Diving

24  

Table 5. 

Median Willingness to Pay by Dive Scenario 

  

Dive Scenario    Model 1  Model 2 

Sally Ward  Median WTP 

95% CI 

$79.99 

($55.86 ‐ $109.80) 

$67.57 

($44.77 ‐ $92.37) 

Wakulla Cavern 

Dive 

Median WTP 

95% CI 

$33.97 

($14.48 ‐ $51.07) 

$22.54 

($3.89 – $40.63) 

Wakulla Cave Dive  Median WTP 

95% CI 

$84.38 

($50.37 ‐ $127.99) 

$69.06 

($34.86 – $106.15)